It uses the Windows Indexer on your SQL server which is pointed to where the file archives are stored.
It doesn't slow down the vault as far as I can tell. When you elect to search by content, it queries the indexed file database outside of the PDM vault database, combines it with the PDM database record, and then returns file results.
I don't see any cons to it......only pros. As long as your users understand how it works, and if you have old scanned documents, it won't find anything on those. Unless you scan them to PDF and run an OCR program on it.
The EPDM manual states
"...Microsoft Indexing Service supplies filters for Microsoft Office files, HTML files, MIME messages, and plain-text files. You can install other document filters for content search in, for example, PDF files."
Do we need "other document filters" for Solidworks & AutoCAD files?
Where do we get these filters?
One other thing to keep in mind, because it uses the Microsoft Indexing Service, and this runs while the server has resources to run it, not all search results will be immediatly available on indexed file.
As far as how to install additional filters, these filters are called iFilters and can be downloaded from various locations. SolidWorks files have their metadata indexed automatically without the need for additional iFilters. The PDF iFilter is installed when you install Adobe Reader on the indexing server.
This iFilter functionality is Microsoft functionality and not SolidWorks functionality. More information can be found on the Microsoft website here: http://msdn.microsoft.com/en-us/library/dd582937(office.11).aspx.
Installing Autodesk products will add ifilters for those file types.
There is a tiff OCR iFilter that comes with Windows Server. http://www.3dvision.com/wordpress/2012/10/22/epdm-works-with-scanned-paper-drawings/
Lots of OCR iFilters out there if you are willing to pay for them.
The MS Office iFilter does a nice job for SW drawings, haven't found a filter that I like for dwg files....
Cons: The index can get very large.
I am wondering if you are using content search and if so how it's working, any barriers to setup?
I setup and configured Content Search but found too many obstacles to make it useful:
- Content search is very limited on file type. You must install the proper “ifiliter” for each file type. Most of our files are autocad dwg and we have many other files types that we would like to search (pdf, inventor,etc) . Finding the right ifilter and keeping them updated is very time consuming.
- Content search creates an index file on the local disk C that can grow to over 40 Gb using all our available disk space. It cannot be relocated. You have to set up indexing separately for each vault, we have 4 vaults.
- We had to modify all our default search cards to show content search.
- Once we finally got everything setup and did a search for a popular text item like CHASSIS we got no results. I know the text exist in many files types.
I plan to retry it after we upgrade our database server sometime later this year.